Annals of Glaciology 58(75pt1) 2017 doi: 10.1017/aog.2017.24 88 © The Author(s) 2017. This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (http://creativecommons. org/licenses/by/4.0/), which permits unrestricted re-use, distribution, and reproduction in any medium, provided the original work is properly cited. Structural provinces of the Ross Ice Shelf, Antarctica Christine M. LEDOUX,1 Christina L. HULBE,2 Martin P. FORBES,2 Ted A. SCAMBOS,3 Karen ALLEY3 1Department of Geology, Portland State University, Portland, OR 97215, USA 2School of Surveying, University of Otago, Dunedin, New Zealand. E-mail: [email protected] 3National Snow and Ice Data Center, University of Colorado Boulder, Boulder, CO, USA ABSTRACT. The surface of the Ross Ice Shelf (RIS) is textured by flow stripes, crevasses and other fea- tures related to ice flow and deformation. Here, moderate resolution optical satellite images are used to map and classify regions of the RIS characterized by different surface textures. Because the textures arise from ice deformation, the map is used to identify structural provinces with common deformation history. We classify four province types: regions associated with large outlet glaciers, shear zones, exten- sion downstream of obstacles and suture zones between provinces with different upstream sources. Adjacent provinces with contrasting histories are in some locations deforming at different rates, suggest- ing that our province map is also an ice fabric map. Structural provinces have more complicated shapes in the part of the ice shelf fed by West Antarctic ice streams than in the part fed by outlet glaciers from the Transantarctic Mountains. The map may be used to infer past variations in stress conditions and flow events that cannot be inferred from flow traces alone. Keywords: crevasses, ice shelves, structural glaciology INTRODUCTION both the ambient stress field and near-field modifications to Linear features observed at the surface of a flowing ice mass that field arising from the fractures themselves. Material prop- represent an integrated history along the flow trajectory. In an erties of the ice may modify its fracture toughness and thus ice shelf, that history includes variation in flow of tributary either promote or suppress propagation (Khazendar and glaciers, conditions at the coastline where the ice went others, 2007; Glasser and others, 2009; Hulbe and others, afloat, and processes within the floating ice (see Fahnestock 2010; Borstad and others, 2012). Most crevasses visible at and others, 2000; Glasser and others, 2009, 2015;Elyand the surface remain relatively short over most of the advective Clark, 2016). Linear features include both along-flow-oriented pathway, and only become long (and thus the failure planes flow stripes and crevasses whose presence and geometry are for large icebergs) in the near-front environment, an effect determined by principal stresses in the ice (e.g. McGrath and that can be related to both material properties and the inter- others, 2012; Colgan and others, 2016). Flow stripes at the action among nearby fractures (cf. Lazzara and others, 1999; surface can be traced back upstream to source regions and, Hulbe and others, 2010; Brunt and others, 2011). together with textures due to crevasses, can be used to identify Our primary interest is in the horizontal expression of frac- regions with common history and to infer deformation history ture propagation as a mixed-mode opening (mode I) and of the ice within each region. sliding (mode II) phenomenon and governed by conditions Surface features record ice shelf history in different ways. within the ice shelf. Once initiated, crevasses are advected Once formed, flow stripes and crevasses are transported in the flow field and may change orientation or shape toward the calving front. Because the flow field is neither spa- depending on the stress field encountered along the way. tially nor temporally uniform, the features experience spa- As passive features, crevasses may be rotated, elongated, tially and temporally varying stress and strain conditions. closed or otherwise modified by the viscous flow of the ice. This gives rise to time-varying deformation that may be inter- They may also be buried by new snow. Where crevasses preted in terms of past variability. For example, streaklines continue to propagate or are reactivated during advection, are surface undulations associated with flow variations in their shapes record stresses experienced along the path and grounded source ice (Glasser and others, 2015). Once perhaps also changes in the ice shelf (MacAyeal and afloat, these flow features become passive tracers and their Lange, 1988; Hulbe and Fahnestock, 2007). deformation relative to the modern flow field has in the The analysis presented here begins with a feature map of case of the Ross Ice Shelf (RIS), been used to infer past flow the RIS created using moderate resolution images of the ice variability (Jezek, 1984; Casassa and others, 1991; shelf surface. Crevasse geometries are classified in a scheme Fahnestock and others, 2000; Hulbe and Fahnestock, 2007). that emphasizes what they reveal about principal stress In the present contribution, we consider flow stripes and orientations. Next, structural provinces (Jackson and Kamb, crevasse patterns together, in order to consider deformation 1997; Herzfeld and others, 2013) within the ice shelf are clas- in the ice shelf more completely. Glacier ice undergoes sified according to pervasive crevasse patterns within regions. brittle fracture at sufficiently large stresses (cf. Rist and Comparison with remotely sensed, present-day strain rates, others, 1999). Fractures nucleate in the subsurface (Nath principal stresses, and ice thickness allow us to discuss ways and Vaughan, 2003) and when stress intensity at a crack in which structural provinces may be used to learn about tip is large enough, the crack will propagate in response to both past and present conditions on the ice shelf. Downloaded from https://www.cambridge.org/core. 29 Sep 2021 at 19:41:04, subject to the Cambridge Core terms of use. LeDoux and others: Structural provinces of the Ross Ice Shelf, Antarctica 89 DATA rate results, we prefer the Landsat-8-derived field for its sharper and more easily managed artifacts. Uncertainty in − Image mosaics the Landsat-8-derived velocity field is 5 m a 1 and the propa- Clear feature geometries that are straightforward to digitize gated uncertainty on the effective strain rate, using gradients − are required for mapping crevasse patterns and regional computed on a 750 m regular grid, is 0.008 a 1. This is surface textures. We are interested in groups of features similar to errors reported for MEaSUREs. The estimated with similar orientations, which produce textures at the ice error does not capture noise in the effective strain rate field. _ surface that persist over tens of km rather than finer, sub- The 1 − σ standard deviation in the ee field downstream of km scale, details. While automated approaches exist (Ely the large outlet glaciers discussed in a later section is − and Clark, 2016), we hand-digitize features with the inten- 0.016 a 1. σ0 ¼ σ ð = Þσ δ tion of developing familiarity with the textures and the mech- Deviatoric stresses are computed as ij ij 1 3 ij ij δ δ σ anical settings in which they are produced. These objectives in which ij is the Kronecker . The stress tensor ij is esti- e make moderate resolution optical images suitable for our mated using the finite strain rates ij and the inverse form of work. the generalized flow law for ice Two image mosaics are the primary data sources for our map: the first epoch of the MODIS (Moderate Resolution σ ¼ e_ð1=nÞ1e_ ð Þ ij B e ij 2 Imaging Spectroradiometer) Mosaic of Antarctica (MOA; Haran and others, 2005) and the Landsat Image Mosaic of in which the exponent n is equal to 3. We use a value of 760 − Antarctica (LIMA; Bindschadler and others, 2008). The first kPa a 1/3 for the temperature-dependent rate factor B, epoch of the MOA is a composite of 260 individual equivalent to a depth average temperature of 247 K (Van MODIS images acquired between 20 November 2003 and der Veen, 1999, p. 16). Principal stress directions and magni- 29 February 2004 with an effective resolution of 125 m. tudes are computed as the eigenvectors and eigenvalues of The LIMA, which has a resolution of 30 m and covers the deviatoric stress tensor. The uniform rate factor is very ° ice shelf region north of about 82 S, is a composite of simple, but our interest is in the eigenvector directions nearly 1100 images acquired by the Landsat ETM+ rather than the values. (Enhanced Thematic Mapper Plus) sensor between 1999 and 2003. Because the MOA composite amplifies shadows, low-relief features like along-flow lineations, Ice surface elevation sagging snow bridges, and regional textures arising from We use the Digital Elevation Model-derived from measure- snow-covered features are readily apparent in the mosaic. ments made by the Geoscience Laser Altimeter (GLAS) We thus use the MOA to make an initial mapping of the (DiMarzio, 2007) to supplement the discussion of ice shelf whole ice shelf surface and turn to the higher resolution provinces. In general, thicker ice floats higher in the ocean LIMA and some individual Landsat 8 scenes, obtained via surface but we do not convert the surface elevation into the USGS EarthExplorer (earthexplorer.usgs.gov), to resolve thickness because we lack knowledge of spatial variation ambiguities regarding boundaries of some textured regions. in snow density and because this is not necessary for the intended use of the data. Velocity, strain rates and stresses METHODS Glaciological strain rates and stresses, computed using remotely sensed surface velocity, are used to support inter- A wealth of surface features are visible in the MOA and LIMA pretation of the feature maps.